<html><head><meta http-equiv="Content-Type" content="text/html charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class="">Hi, Vladimir & Igor,<div class=""><br class=""></div><div class="">I found you decide to throw SkippedException if tieredCompilation doesn’t meet. </div><div class="">I modified my patch to follow this direction. Could you review it? </div><div class=""><div class=""><a href="http://cr.openjdk.java.net/~phh/8207965/webrev.03/" class="">http://cr.openjdk.java.net/~phh/8207965/webrev.03/</a></div><div class=""><br class=""></div><div class="">Besides the inputs from Vladimir, I mainly worked on TestOnSpinWait</div><div class="">It will run both client and server VMs. It will skip c1-check if user only enable compiler2. </div><div class=""><br class=""></div><div class="">Tested using c2-only server and clientVM locally(x86-64+macosx).</div><div class=""><br class=""></div><div class="">Thanks,</div><div class="">—lx</div><div class=""><br class=""></div><div><blockquote type="cite" class=""><div class="">On Aug 1, 2018, at 1:05 PM, Hohensee, Paul <<a href="mailto:hohensee@amazon.com" class="">hohensee@amazon.com</a>> wrote:</div><br class="Apple-interchange-newline"><div class=""><div class="WordSection1" style="page: WordSection1; font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: normal; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px;"><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class="">Webrev also here:<span class="Apple-converted-space"> </span><a href="http://cr.openjdk.java.net/~phh/8207965/webrev.02/" style="color: purple; text-decoration: underline;" class="">http://cr.openjdk.java.net/~phh/8207965/webrev.02/</a><o:p class=""></o:p></div><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div><div style="border-style: solid none none; border-top-width: 1pt; border-top-color: rgb(181, 196, 223); padding: 3pt 0in 0in;" class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><b class=""><span style="font-size: 12pt;" class="">From:<span class="Apple-converted-space"> </span></span></b><span style="font-size: 12pt;" class="">hotspot-compiler-dev <<a href="mailto:hotspot-compiler-dev-bounces@openjdk.java.net" class="">hotspot-compiler-dev-bounces@openjdk.java.net</a>> on behalf of Liu Xin <<a href="mailto:navy.xliu@gmail.com" class="">navy.xliu@gmail.com</a>><br class=""><b class="">Date:<span class="Apple-converted-space"> </span></b>Tuesday, July 31, 2018 at 6:31 PM<br class=""><b class="">To:<span class="Apple-converted-space"> </span></b>Vladimir Kozlov <<a href="mailto:vladimir.kozlov@oracle.com" class="">vladimir.kozlov@oracle.com</a>><br class=""><b class="">Cc:<span class="Apple-converted-space"> </span></b>"<a href="mailto:hotspot-compiler-dev@openjdk.java.net" class="">hotspot-compiler-dev@openjdk.java.net</a>" <<a href="mailto:hotspot-compiler-dev@openjdk.java.net" class="">hotspot-compiler-dev@openjdk.java.net</a>><br class=""><b class="">Subject:<span class="Apple-converted-space"> </span></b>Re: RFR(XS): fix c2-only compilation and some tests<o:p class=""></o:p></span></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class="">Hi, Vladimir, <span class="Apple-converted-space"> </span><o:p class=""></o:p></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class="">thanks for running it and the feedback. <o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class="">I made a modification according your feedback. Could you take a look?<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><a href="https://s3-us-west-2.amazonaws.com/openjdk-webrevs/jdk/c2_only_fix-v3/webrev/index.html" style="color: purple; text-decoration: underline;" class="">https://s3-us-west-2.amazonaws.com/openjdk-webrevs/jdk/c2_only_fix-v3/webrev/index.html</a><o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class="">1. I don't suffer from SA failure. can I pass them all. <o:p class=""></o:p></div></div><div class=""><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""> TEST TOTAL PASS FAIL ERROR<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""> jtreg:test/hotspot/jtreg/serviceability/sa 22 22 0 0<o:p class=""></o:p></div></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class="">ClhsdbFindPC.java seems not to relate to 'TieredCompilation'<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class="">2. I don't quite understand the testcase<o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><span style="font-size: 9.5pt; background-color: white; background-position: initial initial; background-repeat: initial initial;" class="">test/hotspot/jtreg/vmTestbase/jit/tiered/tieredTest.sh</span><o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><span style="font-size: 9.5pt; background-color: white; background-position: initial initial; background-repeat: initial initial;" class=""><br class=""><br class=""></span><o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><span style="font-size: 9.5pt; background-color: white; background-position: initial initial; background-repeat: initial initial;" class="">The java file is empty. it only triggers a shell without any parameter. </span><o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><span style="font-size: 9.5pt; background-color: white; background-position: initial initial; background-repeat: initial initial;" class="">Actually, this test is dummy unless you pass in JTREG="VM_OPTIONS=-XX:+TieredCompilation"</span><o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><span style="font-size: 9.5pt;" class="">how to run it with different VM_OPTIONS. </span><o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><span style="font-size: 9.5pt;" class="">thanks,</span><o:p class=""></o:p></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><span style="font-size: 9.5pt;" class="">--lx</span><o:p class=""></o:p></div></div></div></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><o:p class=""> </o:p></div><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class="">On Mon, Jul 30, 2018 at 10:24 PM, Vladimir Kozlov <<a href="mailto:vladimir.kozlov@oracle.com" target="_blank" style="color: purple; text-decoration: underline;" class="">vladimir.kozlov@oracle.com</a>> wrote:<o:p class=""></o:p></div><blockquote style="border-style: none none none solid; border-left-width: 1pt; border-left-color: rgb(204, 204, 204); padding: 0in 0in 0in 6pt; margin-left: 4.8pt; margin-right: 0in;" class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class="">I hit several issues in testing.<br class=""><br class="">1. Build failures on SPARC because of missing include:<br class=""><br class="">make/hotspot/src/native/dtrace/generateJvmOffsets.cpp<br class="">@@ -40,6 +40,7 @@<br class=""><br class=""> #include <proc_service.h><br class=""> #include "gc/shared/collectedHeap.hpp"<br class="">+#include "memory/heap.hpp"<br class=""> #include "runtime/vmStructs.hpp"<br class=""><br class=""> typedef enum GEN_variant {<br class=""><br class="">2. Several tests failed because they want -XX:+TieredCompilation. Currently there is no check<br class=""><br class="">vmTestbase/jit/tiered/TestDescription.java<br class=""><br class="">test/hotspot/jtreg/vmTestbase/jit/tiered/tieredTest.sh Mon Jul 30 21:48:51 2018 -0700<br class="">@@ -46,6 +46,11 @@<br class=""> exit 0<br class=""> fi<br class=""><br class="">+if grep "TieredCompilation not supported in this VM" $log; then<br class="">+ echo "TEST PASSED: Non-tiered Server VM. The test is useless"<br class="">+ exit 0<br class="">+fi<br class="">+<br class=""> if ! egrep '^[0-9.]+: \[compile level=[0-9]' $log; then<br class=""> if [ "${tiered}" == "on" ]; then<br class=""> echo "TEST FAILED: No PrintTieredEvents output"<br class=""><br class="">compiler/tiered/ConstantGettersTransitionsTest.java (change assert to check) - run with -ea -esa<br class=""><br class="">compiler/onSpinWait/TestOnSpinWait.java (add skipOnTieredCompilation() check to skip C1 test)<br class=""><br class="">There is also closed test failed because it uses C1 flag - need to add -XX:+IgnoreUnrecognizedVMOptions (it is work for us in Oracle).<br class=""><br class="">serviceability/sa/ClhsdbFindPC.java (I don't know how to fix it yet)<span style="color: rgb(136, 136, 136);" class=""><br class=""><br class=""><span class="hoenzb">Vladimir</span></span><o:p class=""></o:p></div><div class=""><div class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class=""><br class=""><br class=""><br class="">On 7/30/18 6:19 PM, Vladimir Kozlov wrote:<o:p class=""></o:p></div><blockquote style="border-style: none none none solid; border-left-width: 1pt; border-left-color: rgb(204, 204, 204); padding: 0in 0in 0in 6pt; margin-left: 4.8pt; margin-right: 0in;" class=""><div style="margin: 0in 0in 0.0001pt; font-size: 11pt; font-family: Calibri, sans-serif;" class="">I started our tier1-3 testing with --with-jvm-features=-compiler1 build.<br class=""><br class="">Regards,<br class="">Vladimir<br class=""><br class="">On 7/30/18 3:03 PM, Liu Xin wrote:<o:p class=""></o:p></div><blockquote style="border-style: none none none solid; border-left-width: 1pt; border-left-color: rgb(204, 204, 204); padding: 0in 0in 0in 6pt; margin-left: 4.8pt; margin-right: 0in;" class=""><p class="MsoNormal" style="margin: 0in 0in 12pt; font-size: 11pt; font-family: Calibri, sans-serif;">hi, list,<br class=""><br class="">We also submitted to the submit repo and ran it successfully.<br class=""><br class="">I am not sure if it runs in the special configuration. This document doesn't say anything about configurations.<br class=""><a href="https://wiki.openjdk.java.net/display/Build/Submit+Repo" target="_blank" style="color: purple; text-decoration: underline;" class="">https://wiki.openjdk.java.net/display/Build/Submit+Repo</a><br class=""><br class="">thanks,<br class="">--lx<br class=""><br class=""><br class="">On Mon, Jul 30, 2018 at 2:41 PM, Liu Xin <<a href="mailto:navy.xliu@gmail.com" target="_blank" style="color: purple; text-decoration: underline;" class="">navy.xliu@gmail.com</a><span class="Apple-converted-space"> </span><mailto:<a href="mailto:navy.xliu@gmail.com" target="_blank" style="color: purple; text-decoration: underline;" class="">navy.xliu@gmail.com</a>>> wrote:<br class=""><br class=""> hi, mail-list,<br class=""><br class=""> Could you kindly review this patch?<br class=""> Because the patch is from Vladimir, we might need another reviewer to approve it.<br class=""><br class=""> Problem: JDK-8207965<br class=""> Webrev:<span class="Apple-converted-space"> </span><a href="http://cr.openjdk.java.net/~phh/8207965/webrev.01/" target="_blank" style="color: purple; text-decoration: underline;" class="">http://cr.openjdk.java.net/~phh/8207965/webrev.01/</a><br class=""> <<a href="http://cr.openjdk.java.net/~phh/8207965/webrev.01/" target="_blank" style="color: purple; text-decoration: underline;" class="">http://cr.openjdk.java.net/~phh/8207965/webrev.01/</a>><br class=""><br class=""> I passed the hotspot-tier1 on x86-64 with the following configuration:<br class=""> --with-debug-level=fastdebug --with-jvm-features=-compiler1,zgc<br class=""><br class=""> thanks,<br class=""> --lx<br class=""><br class=""><br class=""> On Fri, Jul 27, 2018 at 5:27 PM, Liu Xin <<a href="mailto:navy.xliu@gmail.com" target="_blank" style="color: purple; text-decoration: underline;" class="">navy.xliu@gmail.com</a><span class="Apple-converted-space"> </span><mailto:<a href="mailto:navy.xliu@gmail.com" target="_blank" style="color: purple; text-decoration: underline;" class="">navy.xliu@gmail.com</a>>> wrote:<br class=""><br class=""> hi, Vladimir,<br class=""> thank for the patch. I will verify it locally.<br class=""> I will take a look at ZGC.<br class=""><br class=""> thanks,<br class=""> --lx<br class=""><br class=""><br class=""> On Fri, Jul 27, 2018 at 5:19 PM, Vladimir Kozlov <<a href="mailto:vladimir.kozlov@oracle.com" target="_blank" style="color: purple; text-decoration: underline;" class="">vladimir.kozlov@oracle.com</a><br class=""> <mailto:<a href="mailto:vladimir.kozlov@oracle.com" target="_blank" style="color: purple; text-decoration: underline;" class="">vladimir.kozlov@oracle.com</a>>> wrote:<br class=""><br class=""> Hi Liu<br class=""><br class=""> I looked on it again and you are right. Method::_aot_code is defined and used only in<br class=""> tiered case.<br class=""><br class=""> But changes should be next:<br class=""> 1. The assert is needed for case when aot code is used.<br class=""> 2. AOTCompiledMethod::make_entrant() is used only in Tiered (add #ifdef guard).<br class=""> 3. #ifdef DirectivesStack::init() should check INCLUDE_JVMCI<br class=""> 4. Missing changes for ZGC.<br class=""> 5. In TestMeetIncompatibleInterfaceArrays.java skip test with C1 compilation request<br class=""> when Tiered is off.<br class=""><br class=""> Thanks,<br class=""> Vladimir<br class=""><br class=""><br class=""> diff -r 2ce72467c4e8 src/hotspot/share/aot/aotCodeHeap.cpp<br class=""> --- a/src/hotspot/share/aot/aotCodeHeap.cpp<br class=""> +++ b/src/hotspot/share/aot/aotCodeHeap.cpp<br class=""> @@ -714,7 +714,7 @@<br class=""> void AOTCodeHeap::sweep_method(AOTCompiledMethod *aot) {<br class=""> int indexes[] = {aot->method_index()};<br class=""> sweep_dependent_methods(indexes, 1);<br class=""> - vmassert(aot->method()->code() != aot && aot->method()->aot_code() == NULL, "method<br class=""> still active");<br class=""> + vmassert(aot->method()->code() != aot TIERED_ONLY( && aot->method()->aot_code() ==<br class=""> NULL), "method still active");<br class=""> }<br class=""><br class=""><br class=""> diff -r 2ce72467c4e8 src/hotspot/share/aot/aotCompiledMethod.cpp<br class=""> --- a/src/hotspot/share/aot/aotCompiledMethod.cpp<br class=""> +++ b/src/hotspot/share/aot/aotCompiledMethod.cpp<br class=""> @@ -206,6 +206,7 @@<br class=""> return true;<br class=""> }<br class=""><br class=""> +#ifdef TIERED<br class=""> bool AOTCompiledMethod::make_entrant() {<br class=""> assert(!method()->is_old(), "reviving evolved method!");<br class=""> assert(*_state_adr != not_entrant, "%s", method()->has_aot_code() ? "has_aot_code()<br class=""> not cleared" : "caller didn't check has_aot_code()");<br class=""> @@ -240,6 +241,7 @@<br class=""><br class=""> return true;<br class=""> }<br class=""> +#endif // TIERED<br class=""><br class=""> // We don't have full dependencies for AOT methods, so flushing is<br class=""> // more conservative than for nmethods.<br class=""> diff -r 2ce72467c4e8 src/hotspot/share/aot/aotCompiledMethod.hpp<br class=""> --- a/src/hotspot/share/aot/aotCompiledMethod.hpp<br class=""> +++ b/src/hotspot/share/aot/aotCompiledMethod.hpp<br class=""> @@ -194,7 +194,7 @@<br class=""> virtual address verified_entry_point() const { return _code +<br class=""> _meta->verified_entry_offset(); }<br class=""> virtual void log_identity(xmlStream* stream) const;<br class=""> virtual void log_state_change() const;<br class=""> - virtual bool make_entrant();<br class=""> + virtual bool make_entrant() NOT_TIERED({ ShouldNotReachHere(); return false; });<br class=""> virtual bool make_not_entrant() { return make_not_entrant_helper(not_entrant); }<br class=""> virtual bool make_not_used() { return make_not_entrant_helper(not_used); }<br class=""> virtual address entry_point() const { return _code + _meta->entry_offset(); }<br class=""> diff -r 2ce72467c4e8 src/hotspot/share/compiler/compilerDirectives.cpp<br class=""> --- a/src/hotspot/share/compiler/compilerDirectives.cpp<br class=""> +++ b/src/hotspot/share/compiler/compilerDirectives.cpp<br class=""> @@ -442,7 +442,7 @@<br class=""> char str[] = "*.*";<br class=""> const char* error_msg = NULL;<br class=""> _default_directives->add_match(str, error_msg);<br class=""> -#ifdef COMPILER1<br class=""> +#if defined(COMPILER1) || INCLUDE_JVMCI<br class=""> _default_directives->_c1_store->EnableOption = true;<br class=""> #endif<br class=""> #ifdef COMPILER2<br class=""> diff -r 2ce72467c4e8 src/hotspot/share/gc/z/zBarrierSet.cpp<br class=""> --- a/src/hotspot/share/gc/z/zBarrierSet.cpp<br class=""> +++ b/src/hotspot/share/gc/z/zBarrierSet.cpp<br class=""> @@ -22,8 +22,12 @@<br class=""> */<br class=""><br class=""> #include "precompiled.hpp"<br class=""> +#ifdef COMPILER1<br class=""> #include "gc/z/c1/zBarrierSetC1.hpp"<br class=""> +#endif<br class=""> +#ifdef COMPILER2<br class=""> #include "gc/z/c2/zBarrierSetC2.hpp"<br class=""> +#endif<br class=""> #include "gc/z/zBarrierSet.hpp"<br class=""> #include "gc/z/zBarrierSetAssembler.hpp"<br class=""> #include "gc/z/zGlobals.hpp"<br class=""> @@ -33,8 +37,8 @@<br class=""><br class=""> ZBarrierSet::ZBarrierSet() :<br class=""> BarrierSet(make_barrier_set_assembler<ZBarrierSetAssembler>(),<br class=""> - make_barrier_set_c1<ZBarrierSetC1>(),<br class=""> - make_barrier_set_c2<ZBarrierSetC2>(),<br class=""> + COMPILER1_PRESENT( make_barrier_set_c1<ZBarrierSetC1>() )<br class=""> NOT_COMPILER1(NULL),<br class=""> + COMPILER2_PRESENT( make_barrier_set_c2<ZBarrierSetC2>() )<br class=""> NOT_COMPILER2(NULL),<br class=""> BarrierSet::FakeRtti(BarrierSet::ZBarrierSet)) {}<br class=""><br class=""> ZBarrierSetAssembler* ZBarrierSet::assembler() {<br class=""> diff -r 2ce72467c4e8<br class=""> test/hotspot/jtreg/compiler/types/TestMeetIncompatibleInterfaceArrays.java<br class=""> --- a/test/hotspot/jtreg/compiler/types/TestMeetIncompatibleInterfaceArrays.java<br class=""> +++ b/test/hotspot/jtreg/compiler/types/TestMeetIncompatibleInterfaceArrays.java<br class=""> @@ -362,6 +362,12 @@<br class=""> System.out.println((j + 1) + ". invokation of " + baseClassName +<br class=""> i + "ASM.test() [::" +<br class=""> r.getName() + "() should be '" + tier[pass][j]<br class=""> + "' compiled]");<br class=""><br class=""> + // Skip Profiling compilation (C1) when Tiered is disabled.<br class=""> + boolean profile = (level[pass][j] ==<br class=""> CompilerWhiteBoxTest.COMP_LEVEL_FULL_PROFILE);<br class=""> + if (profile && CompilerWhiteBoxTest.skipOnTieredCompilation(false)) {<br class=""> + continue;<br class=""> + }<br class=""> +<br class=""> WB.enqueueMethodForCompilation(r, level[pass][j]);<br class=""><br class=""> try {<br class=""><br class=""> On 7/26/18 1:11 PM, Liu Xin wrote:<br class=""><br class=""> hi, Vladimir,<br class=""><br class=""> Thank you for replying.<br class=""><br class=""> If I disable compiler1, I will run into compiler errors when I build fastdebug. The<br class=""> attachment is the output.<br class=""> Even if ignore the assertion, we still have problem to pass hs-tier1. it will crash<br class=""> when it try to use jvmci.<br class=""><br class=""> Here is my configuration:<br class=""> The existing configuration has been successfully updated in<br class=""> /Users/xxinliu/Devel/openjdk/jdk/build/macosx-x86_64-normal-server-fastdebug<br class=""> using configure arguments '--enable-option-checking=fatal<br class=""> --with-debug-level=fastdebug --with-jtreg=/Users/xxinliu/Devel/jtreg<br class=""> --with-jvm-features=-compiler1'.<br class=""><br class=""> Configuration summary:<br class=""> * Debug level: fastdebug<br class=""> * HS debug level: fastdebug<br class=""> * JVM variants: server<br class=""> * JVM features: server: 'aot cds cmsgc compiler2 dtrace epsilongc g1gc graal jfr<br class=""> jni-check jvmci jvmti management nmt parallelgc serialgc services vm-structs'<br class=""> * OpenJDK target: OS: macosx, CPU architecture: x86, address length: 64<br class=""> * Version string: 12-internal+0-adhoc.xxinliu.jdk (12-internal)<br class=""><br class=""> thanks,<br class=""> --lx<br class=""><br class=""> On Thu, Jul 26, 2018 at 12:48 PM, Vladimir Kozlov <<a href="mailto:vladimir.kozlov@oracle.com" target="_blank" style="color: purple; text-decoration: underline;" class="">vladimir.kozlov@oracle.com</a><br class=""> <mailto:<a href="mailto:vladimir.kozlov@oracle.com" target="_blank" style="color: purple; text-decoration: underline;" class="">vladimir.kozlov@oracle.com</a>> <mailto:<a href="mailto:vladimir.kozlov@oracle.com" target="_blank" style="color: purple; text-decoration: underline;" class="">vladimir.kozlov@oracle.com</a><br class=""> <mailto:<a href="mailto:vladimir.kozlov@oracle.com" target="_blank" style="color: purple; text-decoration: underline;" class="">vladimir.kozlov@oracle.com</a>>>> wrote:<br class=""><br class=""> This does not seems right. What failure you got in aotCodeHeap.cpp and<br class=""> aotCompiledMethod.cpp?<br class=""><br class=""> Thanks,<br class=""> Vladimir<br class=""><br class=""> On 7/26/18 10:16 AM, Liu Xin wrote:<br class=""><br class=""> Hi, hotspot community,<br class=""><br class=""><br class=""> Please review the small change to getc2-onlybuild to work.<br class=""><br class=""> Bug:<a href="https://bugs.openjdk.java.net/browse/JDK-8207965" target="_blank" style="color: purple; text-decoration: underline;" class="">https://bugs.openjdk.java.net/browse/JDK-8207965</a><br class=""> <<a href="https://bugs.openjdk.java.net/browse/JDK-8207965" target="_blank" style="color: purple; text-decoration: underline;" class="">https://bugs.openjdk.java.net/browse/JDK-8207965</a>><br class=""> <<a href="https://bugs.openjdk.java.net/browse/JDK-8207965" target="_blank" style="color: purple; text-decoration: underline;" class="">https://bugs.openjdk.java.net/browse/JDK-8207965</a><br class=""> <<a href="https://bugs.openjdk.java.net/browse/JDK-8207965" target="_blank" style="color: purple; text-decoration: underline;" class="">https://bugs.openjdk.java.net/browse/JDK-8207965</a>>><br class=""><br class=""> Webrev:<span class="Apple-converted-space"> </span><a href="http://cr.openjdk.java.net/~phh/8207965/webrev.00/" target="_blank" style="color: purple; text-decoration: underline;" class="">http://cr.openjdk.java.net/~phh/8207965/webrev.00/</a><br class=""> <<a href="http://cr.openjdk.java.net/~phh/8207965/webrev.00/" target="_blank" style="color: purple; text-decoration: underline;" class="">http://cr.openjdk.java.net/~phh/8207965/webrev.00/</a>><br class=""> <<a href="http://cr.openjdk.java.net/%7Ephh/8207965/webrev.00/" target="_blank" style="color: purple; text-decoration: underline;" class="">http://cr.openjdk.java.net/%7Ephh/8207965/webrev.00/</a><br class=""> <<a href="http://cr.openjdk.java.net/%7Ephh/8207965/webrev.00/" target="_blank" style="color: purple; text-decoration: underline;" class="">http://cr.openjdk.java.net/%7Ephh/8207965/webrev.00/</a>>><br class=""> <<a href="http://cr.openjdk.java.net/%7Ephh/8207965/webrev.00/" target="_blank" style="color: purple; text-decoration: underline;" class="">http://cr.openjdk.java.net/%7Ephh/8207965/webrev.00/</a><br class=""> <<a href="http://cr.openjdk.java.net/%7Ephh/8207965/webrev.00/" target="_blank" style="color: purple; text-decoration: underline;" class="">http://cr.openjdk.java.net/%7Ephh/8207965/webrev.00/</a>><br class=""><br class=""> <<a href="http://cr.openjdk.java.net/%7Ephh/8207965/webrev.00/" target="_blank" style="color: purple; text-decoration: underline;" class="">http://cr.openjdk.java.net/%7Ephh/8207965/webrev.00/</a><br class=""> <<a href="http://cr.openjdk.java.net/%7Ephh/8207965/webrev.00/" target="_blank" style="color: purple; text-decoration: underline;" class="">http://cr.openjdk.java.net/%7Ephh/8207965/webrev.00/</a>>>><br class=""><br class=""> It will meet 2 compiler errors if you configure project with “./configure<br class=""> --with-jvm-features=-compiler1<br class=""> --enable-option-checking=fatal --with-debug-level=fastdebugâ€.<br class=""><br class=""> the configure disable c1. Furthermore, 12 failures in hs-tier1 if we<br class=""> havec2-only.<br class=""><br class=""><br class=""> C2-onlybuilt met the following errors when I ran hs-tier1:<br class=""><br class=""> compiler/aot/cli/jaotc/CompileClassWithDebugTest.java: check that jaotc can<br class=""> compile a class with a --debug flag<br class=""> compiler/jvmci/compilerToVM/IsCompilableTest.java:<br class=""> compiler/jvmci/events/JvmciNotifyBootstrapFinishedEventTest.java:<br class=""> compiler/jvmci/events/JvmciNotifyInstallEventTest.java:<br class=""> compiler/jvmci/<a href="http://jdk.vm.ci/" style="color: purple; text-decoration: underline;" class="">jdk.vm.ci</a>.code.test/src/jdk/vm/ci/code/test/DataPatchTest.java:<br class=""> compiler/jvmci/jdk.vm.ci.code.test/src/jdk/vm/ci/code/test/MaxOopMapStackOffsetTest.java:<br class=""> compiler/jvmci/<a href="http://jdk.vm.ci/" style="color: purple; text-decoration: underline;" class="">jdk.vm.ci</a>.code.test/src/jdk/vm/ci/code/test/NativeCallTest.java:<br class=""> compiler/jvmci/jdk.vm.ci.code.test/src/jdk/vm/ci/code/test/SimpleCodeInstallationTest.java:<br class=""> compiler/jvmci/jdk.vm.ci.code.test/src/jdk/vm/ci/code/test/SimpleDebugInfoTest.java:<br class=""> compiler/jvmci/jdk.vm.ci.code.test/src/jdk/vm/ci/code/test/VirtualObjectDebugInfoTest.java:<br class=""> compiler/types/TestMeetIncompatibleInterfaceArrays.java: C2 can not handle<br class=""> returns with incompatible interface<br class=""> arrays<br class=""><br class=""> Most due to bug similar withJDK-8145331<br class=""> <<a href="https://bugs.openjdk.java.net/browse/JDK-8145331" target="_blank" style="color: purple; text-decoration: underline;" class="">https://bugs.openjdk.java.net/browse/JDK-8145331</a><br class=""> <<a href="https://bugs.openjdk.java.net/browse/JDK-8145331" target="_blank" style="color: purple; text-decoration: underline;" class="">https://bugs.openjdk.java.net/browse/JDK-8145331</a>><br class=""> <<a href="https://bugs.openjdk.java.net/browse/JDK-8145331" target="_blank" style="color: purple; text-decoration: underline;" class="">https://bugs.openjdk.java.net/browse/JDK-8145331</a><br class=""> <<a href="https://bugs.openjdk.java.net/browse/JDK-8145331" target="_blank" style="color: purple; text-decoration: underline;" class="">https://bugs.openjdk.java.net/browse/JDK-8145331</a>>>>.<br class=""><br class=""><br class=""> CompilerDirectives::get_for(AbstractCompiler *comp) returns _/c1/_store<br class=""><br class=""> if jvmci() enabled, but hotspot won’t enable it if COMPILER1 is<br class=""> disabled. As a result, hotspot messes up<br class=""> refcounts of _c1_store. This webrev fixes 11 out of 12 cases.<br class=""><br class=""><br class=""> The last test (compiler/types/TestMeetIncompatibleInterfaceArrays.java)<br class=""> makes use of WhiteBox to check tier<br class=""> levels. I can’t fix it easily. Shall I try to fix it in this webrev? If no,<br class=""> we can file a new bug to fix the<br class=""> test. Does anyone have any idea how to fix it?<br class=""><br class=""><br class=""> thanks,<br class=""><br class=""> --lx</p></blockquote></blockquote></div></div></blockquote></div></div></div></div></blockquote></div><br class=""></div></body></html>